Requests versus incidents, and finding problems

Why service requests need their own workflows instead of a low-priority slot in the incident queue, what does not belong in the request practice, and the two directions problems are identified from.

Lesson 3 of 12 in objective 7. Seven ITIL practices in detail, part of ITIL 4 Foundation.

Where problems come from — and only one source is reactive. Problem identification contains Incident records and trends (reactive: the pattern behind repeated failures), Monitoring data (proactive: state that predicts failure), Supplier and vendor notifications (proactive: a defect announced before it bites), Testing, projects and reviews (proactive: found before anything is live). Problem identification Incident records and trends reactive: the pattern behind repeated failures Monitoring data proactive: state that predicts failure Supplier and vendor notifications proactive: a defect announced before it bites Testing, projects and reviews proactive: found before anything is live
Where problems come from — and only one source is reactive.

Requests do not belong in the incident queue

Putting service requests into the incident queue as low-priority incidents is a tempting simplification and the exam expects you to be able to argue against it precisely. The strongest objection is not that it distorts the incident statistics, true though that is. It is that requests are agreed, repeatable parts of normal service delivery, so they need their own PRE-DEFINED FULFILMENT WORKFLOWS — and incident handling, which is built to restore something that broke unexpectedly, provides nothing of the kind.

That is also why the two practices measure different things. An incident is judged on how fast normal service came back; a request is judged against the fulfilment target agreed for that kind of request. One queue cannot serve both without losing one of the measures.

What follows when a request is filed as a low-priority incident. In order: Logged as a low-priority incident (the tempting simplification — one queue for everything), then It meets a workflow built for failures (that queue is there to restore what broke unexpectedly), then No pre-defined fulfilment workflow (the strongest objection — agreed, repeatable work with nowhere to run), then Judged on how fast service came back (not against the fulfilment target agreed for that kind of request), then The incident figures are distorted (true, but not the objection to lead with). Logged as a low-priority incident the tempting simplification — one queue for everything It meets a workflow built for failures that queue is there to restore what broke unexpectedly No pre-defined fulfilment workflow the strongest objection — agreed, repeatable work with nowhere to run Judged on how fast service came back not against the fulfilment target agreed for that kind of request The incident figures are distorted true, but not the objection to lead with
What follows when a request is filed as a low-priority incident.

What is not a service request

Requests cover the normal, agreed things users ask for: access, information, a standard item, advice, a routine provisioning. What does not belong is anything that reports a failure. A user saying a service they rely on has become noticeably slower than usual is reporting degraded quality, which is an incident by definition, and it is the option planted in these questions precisely because "the user asked us something" makes it look like a request.

The test is what triggered the contact rather than the form of words. Something broke or degraded, unexpectedly, is an incident. Something normal and agreed was asked for is a request.

Sort the contact by what triggered it, not by how it was worded. One test, and exactly one way out of it. The test is: What triggered the contact? (the form of words is not the test). Something broke or degraded unexpectedly leads to An incident (noticeably slower than usual counts, however politely it is asked); Something normal and agreed was asked for leads to A service request (access, information, a standard item, advice, routine provisioning). What triggered the contact? the form of words is not the test Something broke or degraded unexpectedly Something normal and agreed was asked for An incident noticeably slower than usual counts, however politely it is asked A service request access, information, a standard item, advice, routine provisioning
Sort the contact by what triggered it, not by how it was worded.

How problems come to be identified

Problem identification is the first phase of problem management, and the answer the exam wants covers both directions. Reactively, by analysing incident records and looking for trends — an analyst reviewing several months of incidents and noticing a repeating pattern of similar failures that were each resolved individually is doing problem identification. Proactively, by examining other information — monitoring data, supplier notifications, testing and project findings — which can surface a problem BEFORE any incident has occurred at all.

An answer that describes only the reactive half is incomplete, and it is the most common wrong option here because it matches how most organisations actually behave. The definition of a problem includes potential causes, and this phase is where that word earns its place.

One term to have ready: a known error is an analysed problem that still has no permanent fix. It is not a problem with a workaround, not a problem that has been closed, and not an incident of any kind.

Worth carrying in

Pre-defined fulfilment workflow
What a request needs and an incident queue cannot provide.
Degraded quality
An incident, even when the user phrases it as a question.
Problem identification
First phase. Reactive from incident trends, proactive from other data.
Proactive identification
Monitoring, supplier notifications, testing — before any incident.
Known error
An analysed problem with no permanent fix yet.

What the exam does with this

Objective
7. Seven ITIL practices in detail
Share of the exam
47.5% (the whole objective)
Questions in this lesson
5
Signed for by a person
0

Partly checked. None of the 5 questions here has been read against the cited source by a person. 5 questions have been checked against their cited clause by an automated pass — which is not the same thing, and is not a signature.

Only questions a person has signed for are used in mock exams here. That is the whole difference between the two kinds of checking above.

How these questions are written — where each question comes from, what the verification ledger records, and what happens when one is found wrong.

Drill this lesson

A lesson is one sitting: the trainer draws a short run from these questions alone and spaces the ones you get wrong.

Practise Requests versus incidents, and finding problems

Questions in this lesson

Practise Requests versus incidents, and finding problems

The rest of objective 7